38 INTRODUCTIONsuddenly appear like an inventive mutation, but very slowlyevolve as their usefrdness is demonstrated by practice. The shaft is very rarely perforated at the base for theattachment of a Une i; it is cyUndrical (later flat) in formadapted to the capture of large fish in streams. The harpoonsmay possibly have been projected by means of the so-caUedpropulseurs or dart throwers, which resemble the Eskimo andAustraUan implements of to-day. Amidst the clash of opinion as to the exact use and methodof use of these weapons, my conclusion, admitttedly incapableof absolute proof, holds that the PalaeoUthic fisher owes to thehunter the inception of the chief weapon of his equipment, theSpear-Harpoon. Paul Broca's dictum 2 that Man hunted before he fishedseems, perhaps, despite DaU's excavation of Eskimo dtbris,'^to be bome out by Troglodyte records both positive andnegative. The Gorge or bait-holder was employed by thehunter (according to some) even earUer than by the fisher.Gorges have been from time immemorial and stiU are in voguein the Untersee for the capture of marine birds, as is the caseto-day with the Eskimos of Norton Sound. From the chronicles of Rau, H. PhiUps, and others can bebuilt a Table of Generations, or the story of how the HuntingSpear begat the Fishing Spear, which begat the HarpoonunilateraUy barbed, which in turn begat the Harpoon bi-lateraUy barbed, until about the tenth or twentieth generation—one is appalled at the amount of Succession Duty which such • H. J. Osborne (op. cii., p. 385 fi.) states that, with the exception of onehalf-finished hole in a Harpoon from La Madelaine, the side hole for theattachment of the thong to the Harpoon does not appear in the French Magda¬lenian Harpoon, although in those from Cantabria it is nearly always present.The AziUan weapon usuaUy bears a hole. 2 The Troglodytes of ihe Vhere Valley, Smithsonian Report, 1872, p. 95. ' In Contributions io North American Ethnology, 1877, i. p. 43, DaU statesthat the d/bris of the heaps show tolerably uniform division into three stages,characterised by the food which formed the staple of subsistence and by theweapons for obtaining as well as the utensUs for preparing the food. Thestages are: ist. The Littoral period, represented by the Echinus layer;2nd, The Fishing period, represented by the Fish-bone layer; 3rd, TheHunting period, represented by the MammaUan layer. This antecedenceof fishing before hunting, if DaU be correct, was, I imagine, caused probablyby local or climatic conditions in the Arctic Circle ; it is not the general ruleelsewhere.